On-line services are critical to the brand experience and to fees and profits. So why are they increasingly failing? Because IT and customer workflows are increasingly complex.
Virtual compute infrastructures combined with rapid application delivery combined with ever more dynamic and real-time business events has created a perfect storm for IT operations managers. While they weather a lot of storms successfully, the recent outages
at NatWest and RBS painfully show the costs of these operations storms. The true cost of the outages likely will include extended opening hours for Saturday and Sunday, the financial ombudsman and, no doubt, customer compensation.
Operations has gone web-scale and IT has yet to catch up. In the mainframe and distributed compute days, applications were tethered to physical machines and throwing more hardware and software testing at the problem reduced the failure risk to acceptable levels.
Today however, VM-sprawl, agile applications, and mobile access are causing inter-dependencies between IT components to skyrocket. Simply put, interdependencies are scaling faster than IT. Worse yet, the complexity is now beyond the human ability to see
or manage it.
To get control of this complexity and head-off the risk of failure, service providers have architected their systems for web-scale. Enterprise IT shops need to follow-suit. WebOps thinking will cause a rethink of current IT architectures to more standard,
simpler and scalable levels of abstraction. This is nothing new in the history of enterprise IT. We did it going from mainframes to distributed computing. However, as usual, this time it's happening faster with bigger consequences.
Whatever 'WebOps' means to your IT operations, it means scale, consolidation and reliability to your CIO. We can call it whatever we want but it's time for operations to rethink their approach to automation.